notch776/law_rag_system

A corporate law RAG system with innovative retrieval and contextual strategies

17
/ 100
Experimental

Implements hybrid retrieval combining vector search (Elasticsearch with IK tokenization), BM25, and rule-based pattern matching fused via Reciprocal Rank Fusion, with BGE-reranker-large post-ranking to handle complex multi-document legal queries. Deploys a tiered context architecture using Redis for short-term dialogue caching and Neo4j graph storage for long-term history, while threshold-based context injection controls token consumption and latency. Built on FastAPI/Vue3 with WebSocket support for seamless AI-to-human agent handoff and integrates Alibaba's Qwen-Max LLM via DashScope API.

No commits in the last 6 months.

No License Stale 6m No Package No Dependents
Maintenance 2 / 25
Adoption 8 / 25
Maturity 7 / 25
Community 0 / 25

How are scores calculated?

Stars

62

Forks

Language

Python

License

Last pushed

Sep 28, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/rag/notch776/law_rag_system"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.